Use of neural network language models to study the demand for professional competencies of higher education in the labor market

The paper presents the results of the development of methods for comparing higher education programs and the needs of the labor market. These methods were developed by the authors as part of the implementation of an analytical platform for automated monitoring of the labor market and intelligent analysis of personnel against university specialties. Texts of titles, levels of development and indicators of IT-profile competence achievement, as well as vector language models with various architectures and training text bodies were used for the analysis. The results are mutually compared and the stability of the methods of matching educational competencies and the needs of the labor market is shown.